nVidia – CPU Maker?

Rumors abound concerning the possible entry of nVidia into the CPU market. Though now is more of a tough time to break new things than any in recent memory, I am sure it would be a great idea.

In any market, a third entry makes the other two perk up and get a lot less comfortable than when the dance was only two. It’s much harder to predict how things will go, and personal experience with some Cyrix CPUs in the ’90s tells me that both Cyrix and later, IDT, pushed both Intel and AMD during that time. Perhaps the push was not technologically, but pricing dropped quite a bit during that time, and the customers were benefitted at every level.

Ars Technica tells more of the story -

It’s the NVIDIA rumor that won’t die: no, not the one where the GPU maker buys tiny VIA, but the other one, where it jumps feet-first into competition with both Intel and AMD by producing an x86 processor of its own. The idea has cropped up again in an analyst note from Doug Freedman of Broadpoint AmTech, in which Freedman claims that NVIDIA has been hiring former Transmeta engineers to work on a secret x86 processor that will appear sooner rather than later. In the note seen by the EE Times, Freedman emphasizes that NVIDIA not will take on Intel’s Core i5/i7 lineup. Rather, the company’s plan is to it attack the mid-range to low-end market, possibly competing with AMD in the value segment.

Before unpacking the rumor, let’s lay out the full rationale for the “NVIDIA goes x86, competes with Intel head-on” idea.

First and foremost among the reasons cited for NVIDIA’s alleged plans is the fact that we’re about to make another turn on Sutherland’s wheel of reincarnation, where graphics functionality will move back onto the CPU die. When this happens, NVIDIA’s lucrative integrated graphics processor business is, of course, toast, which has to be one reason why the company went ahead and euthanized it a bit early.

Not that anyone is certain about that last thought, nVidia seems to be lagging a bit, but there are plenty of nVidia products in the channel, and Larrabee from Intel seems to be at least 6 – 9 months of more vaporware.

This turn of the wheel is interesting because it’s not just that the integrated graphics processors (IGP) will move from the chipset to the CPU, but retain their same basic degree of specialization. Rather, both the integrated GPUs that go onto the CPU die and the more discrete models are essentially becoming many-core, general-purpose processors (albeit specialized for multithreaded throughput and floating-point). This makes the situation even more ominous for NVIDIA, because not only will Intel and AMD take their IGP market from them, but even in the discrete GPU market NVIDIA will also end up with a generalized processor that competes with x86.

The ultimate point is that in both the discrete and integrated GPU markets, NVIDIA is already destined to compete directly with Intel and AMD, which means that an actual x86 product from NVIDIA means only that NVIDIA has decided to fight x86 with x86, as opposed to fighting it with some non-x86 architecture

This comes from past days, when non-x86 processors used for any specialized purposes failed miserably. In the ’90s, there were motherboards that featured Intel 486 or Pentium CPUs, along with a secondary processor (many times an Intel i860 or i960), with grandiose claims of how the dual processor arrangement would revolutionize computing. Well, it did, but not the non-identical processing that those times were speaking about. If you remember, by now x86 architecture was to have been dust, as RISC processors, running at hundreds of GHz by now, were to have taken over.

The rest of the article gives a few observations, but no conclusions, and it really is something only a few inside nVidia must know about. However, just as in the graphics market, a strong third would make a big difference, and I’d certainly try an nVidia CPU, as I do have that experience with the little pin compatible Cyrix 486DLC that replaced the 386-33 on my motherboard, and nearly doubled performance with no other changes. Naysayers will speak of how the floating point performance was no better than an Intel 386, but back then only math nerds and accountants knew the difference – nVidia, so completely knowledgeable about FPU – style calculations would hardly make that same mistake. If anything, they might go overboard on the FPU area, giving us a chip with Core2 integer capabilities, and i7 floating point power.

§

⌘

⌘

perhaps this will become nVidia computing, instead of just graphics…

•

•